Ethical issues linked to diagnosis, healthcare and research
2016: Ethical issues linked to the changing definitions/use of terms related to Alzheimer’s disease
In the next part of this discussion paper, we will reflect on the ethical implications of the changing definitions of AD for matters related to people’s health, care and wellbeing, in the context of:
- Treatment and health/social care
In this sub-section on diagnosis, we focus on prodromal AD (IWG) and MCI due to AD (NIA-AA), as pre-clinical AD is currently a research classification and not a clinical diagnosis. We briefly refer to preclinical AD at the end of the sub-section in relation to issues surrounding information about risk status.
Introduction of the new terminology and definitions of AD in the context of diagnosis.
The new AD terminology is gradually being adopted in the clinical setting. However, Morris et al. (2011) describe a 17-year gap for research evidence to reach clinical practice. Moreover, the preclinical research terminologies are still in need of validation for clinical use. It may therefore be some time before diagnoses of prodromal AD (IWG) or MCI due to AD (NIA-AA) become common in clinical settings across Europe. Meanwhile, old diagnostic criteria specifically for AD, such as the NINCDS-ADRDA (McKhann et al. 1984) described earlier or more global classification tools such as the ICD 10 or ones which are specific to mental disorders and diseases such as the the Diagnostic and Statistical Manual (DSM) for Mental Disorders, are most commonly used. Prior to a diagnosis of AD dementia, people with cognitive impairment and little functional impact associated with this are typically diagnosed with Mild Cognitive Impairment (MCI) (Petersen 2004). However, the connection between research and healthcare should not be underestimated. Many centres for the diagnosis of AD, such as memory clinics, take part in research. Thus, the timelag between research and clinical practice may sometimes be less than in other areas of healthcare.
Until recently, MCI has been considered as a condition which does not necessarily lead to dementia, with many people reverting back to normal cognitive functioning or remaining stable over time (Petersen et al. 2014). In the new AD model, however, the sub-category known as MCI due to AD (NIA-AA), which is more specific than MCI in general, is considered as a stage of AD. According to Dubois et al.:
“The proposed conceptual shift is to consider a patient previously diagnosed as having MCI (ie, with an amnestic syndrome of the hippocampal type and with biomarker evidence positive for brain amyloidosis) to be no longer at risk for developing AD dementia, but to recognise that they already have AD at a prodomal stage with an inevitable progression to AD dementia over time.” (Dubois et al. 2010, 1123)
The inevitable progression to AD dementia mentioned above is an important piece of information, but at present conjecture, as research in this area is not yet conclusive. The classifications are relatively new and the inevitable progession over time therefore remains to be proven. Studies are needed which span lengthy periods of time during which people are likely to progress along the AD continuum, thus making it possible to determine whether progression to AD dementia is inevitable.
The timeliness of diagnosis
In recent years, the importance of the timeliness of diagnosis has been emphasised (e.g. see Scottish Government, 2013). Whilst the terms “timely” and “early” diagnosis are often used interchangeably, “timely” refers to a diagnosis which is made at the right time for a particular person, whereas “early” focuses on a diagnosis which is made as early as possible (i.e. in the chronological sense) (Dhedi et al., 2014). According to Woods et al. (2003, p.321) timely diagnoses “prevent crises, facilitate adjustment and provide access to treatments and support”. In keeping with a person-centred approach, timely diagnosis is not linked to a particular disease stage but to benefit to the individual patient (Brayne, 2012 in Dhedi et al. 2014). Although, there is, as yet, no treatment or post-diagnostic support for people with prodromal AD (IWG) or MCI due to AD (NIA-AA), some people may find it beneficial to know their AD status so as to plan for the future. On the other hand, this may in some cases involve planning for a future which does not happen and result in unnecessary distress and lifestyle changes. A potential and significant benefit at the level of society, however, is that through the identification of these subgroups, it may eventually be possible to achieve targeted therapy. This involves identifying the ‘at risk’ person and tailoring treatment to reduce modifiable risks, enhance resilience and potentially modify the course of the disease through specific drug interventions.
Potential harm must also be considered (e.g. stigma, the consequences of incorrect diagnosis and various emotional and psychological reactions). There is as yet insufficient evidence about harm linked to a diagnosis of prodromal AD (Dubois et al. 2016). However, it must be considered whether certain diagnostic procedures such as lumbar punctures, which are potentially burdensome and involve some degree of risk, are justifiable in the absence of curative treatment. In keeping with the principle of respect for autonomy, people should be given the opportunity to decide whether or not they would like to be informed of their diagnosis (Molinuevo and Rami 2013). The right not to know is equally important (Marzanski 2000) and for this to be a genuine choice, people need to understand fully what such a diagnosis means. Potential confusion and conflicting messages surrounding AD-related terms (see next sub-section) may lead to uncertainty about what a person would or would not want to know.
Timely diagnosis and subsequent treatment should be based on sufficient scientific evidence and ethical reflection. However, “therapeutic nihilism” often interferes with timely diagnosis of AD. This involves the belief held by some healthcare professionals that it is pointless to diagnose AD (i.e. AD dementia) as there is no treatment, a risk of stigma and as they feel they have nothing to offer (van Hout et al. 2000; Vernooij-Dassen, Moniz-Cook et al. 2005). The likelihood of a yet earlier diagnosis being perceived by professionals as beneficial is likely to be even lower. Another key barrier is confidence in the diagnosis and in the ability to diagnose (Koch and Iliffe 2010). The new definitions could help clarify whether a person has AD dementia or another dementia-related condition. Ongoing education of healthcare professionals is therefore vitally important to ensure that the new diagnostic criteria and AD model do not increase uncertainty about the value of diagnosis but rather empower healthcare professionals by contributing towards greater diagnostic precision and eventually to the possibility to offer more targeted treatment and care.
Communication of the new definitions of AD in the context of diagnosis
Respect for autonomy involves giving people the opportunity to make voluntary decisions (e.g. about whether they want to be informed of a diagnosis and/or about their lives following diagnosis) based on a full understanding of the facts. Doctors often use euphemisms or non-medical terms to help patients understand diagnoses of dementia (Cody et al. 2002; Downs et al. 2002). However, the use of a mixture of terms related to AD (e.g. reflecting AD as a form of dementia or as a continuum covering both pre- and post dementia stages) may sometimes contribute towards misunderstandings. In addition, lay people may find some of the new terms confusing. “MCI due to AD”, for example, could be interpreted as referring to a form of MCI which is due to AD or as implying that MCI (i.e. in general) is due to AD. Similarly, prodromal is not an everyday term and most people will not know what it means. Sometimes, the term “MCI/prodromal AD” is used and it is not clear whether the “AD” is linked solely to prodromal or also to MCI (Jessen et al. 2014) and AD continues to be described in numerous publications as a form of dementia.
There is also a risk of some healthcare professionals interpreting or framing a diagnosis of prodromal AD (IWG) or MCI due to AD (NIA-AA) as a very mild form of AD dementia (e.g. in an attempt to make the diagnosis meaningful to their patients). In this way, people may be exposed to potentially conflicting messages from different reliable sources. This may cause distress and undermine trust in healthcare professionals. Doctors may need more time than is typically available, as well as the appropriate skills and training, first to grasp and secondly to make the new definitions of AD understandable to their patients. This discussion paper will hopefully serve as a starting point for clarification and for dialogue with different target groups but it is essential to understand how people from different groups in society make sense of the information they receive (please see section on Media/Public Awareness).
In keeping with the principle of social justice, which involves the fair distribution of resources, opportunities and potential burden and benefits within society, if there are benefits to be had from being diagnosed with AD, then everyone should have the opportunity to receive such a diagnosis. Benefits for people diagnosed with prodromal AD or MCI due to AD (where this is already occurring), in terms of care, currently seem fairly limited and the burden of the tests (e.g. worry, discomfort or some degree of pain, perceived risk of complications) might for some individuals insufficiently outweigh any perceived benefits. This and the cost or lack of access to lumbar punctures and PET scans may lead to inequity in access to diagnosis.
Research is currently being carried out to develop a “low cost” battery of measures (e.g. simple tests and games) which would enable a cost-efficient, more holistic and non-invasive early detection of people at risk of developing dementia (PredictND 2016). Ideally, such methods might eventually contribute towards diagnosis in cases where access to the necessary diagnostic tools is unfortunately lacking or where people do not wish to undergo the tests proposed. Nevertheless, the tests currently being developed are aimed at earlier detection of cognitive decline rather than actual diagnosis and such approaches to diagnosis might be considered second best if other methods are not available.
Preclinical AD and diagnosis
So far, we have focused on diagnosis in relation to prodromal AD (IWG) and MCI due to AD (NIA-AA), as pre-clinical AD is currently a research classification and not a clinical diagnosis. However, the term ‘diagnosis’ is sometimes used in connection with preclinical AD, perhaps because of the blurred boundaries between research and clinical practice mentioned earlier. The value of labelling asymptomatic-at-risk people as having a disease, when many of them may never develop any symptoms, has nevertheless been questioned (Giaccone et al. 2011; Covinsky 2011).
“A person who has pathological changes that are unlikely to cause symptoms is better classified as normal than as having preclinical disease. Diagnosing preclinical Alzheimer’s disease in someone destined to never have cognitive problems should be viewed as a misdiagnosis.” (Covinsky 2011)
On the other hand, if a risk has been detected, people should have the opportunity to be informed of it in keeping with the principle of respect for autonomy. Such information would enable them to decide whether or not to take advantage of any preventive treatment that might eventually become available and/or to participate in preventive clinical research, also deciding what risks, if any, they might be willing to take in connection with these two possibilities. Nevertheless, more research is needed into the psychological, emotional and social impact of receiving information about AD at all stages along the continuum (linked to diagnosis for prodromal AD or MCI due to AD and for AD dementia, and about risk status in relation to preclinical AD). Adequate pre-diagnostic counselling and post-diagnostic support should be provided to help maintain the equal standing of individuals within the civil, social and political community.
Treatment and health/social care
The changing definitions of Alzheimer’s disease (AD), if adopted clinically, will have an inevitable impact on the nature and extent of health and social care services provided for people with AD pathology. Although the purpose of changing definitions has been driven by the field of research, it is acknowledged that a related purpose of the lexicon is to impact on clinical communities (Dubois et al. 2010). Their adoption will depend on the perceived utility of the definitions from the perspective of practitioners, as well as from the perspective of people who receive health and social care. There are implications linked to the impact that changing terminologies will have on consumer relationships with healthcare services, related to both pharmaceutical and psychosocial treatment options. In addition, the adoption of the prodromal AD (IWG) or MCI due to AD (NIA-AA) classification by clinicians will open a discussion regarding the eligibility for treatment and access to health and social care services for people with this category of diagnosis.
Health and social care interventions
The prodromal AD (IWG) and MCI due to AD (NIA-AA) classifications, evidence of pathology and early clinical symptoms, loosely align to the current use of terminology such as MCI in practice (albeit to a sub-category of MCI). Current treatment options for MCI are typically limited to monitoring the progression of symptoms to detect clinically significant decline (e.g. UK-based NICE 2006 guidance). However, there is evidence that individuals with MCI could benefit from receiving targeted interventions. For example, there is evidence that cognitive training can reduce memory impairment in MCI (Moro et al. 2012). Furthermore, there is evidence that potential significant benefit for this group can be achieved by lifestyle changes (Palmer et al. 2010). Active informational interventions promoting the benefits of exercise and a healthy diet could be worthwhile for this group. Still, it should be considered that, even though cognitive training and lifestyle changes have shown a benefit to targeted groups in the population with specific characteristics, they may not be effective for each individual.
If AD is accepted as a continuum, and prodromal AD (IWG) or MCI due to AD (NIA-AA) as a precursor to AD dementia, the question arises whether existing interventions recommended for people with AD dementia should be extended to this group. There is increasing evidence for the efficacy of cognitive-based interventions for people with dementia. A systematic review of 15 randomised controlled trials (RCTs) of cognitive stimulation therapy, and meta-analysis of results for 718 participants confirm the benefits of this approach for cognition and quality of life in people living with dementia (Woods et al. 2012). Policy initiatives in the UK, for example, recommend that all people with mild/moderate dementia of all types be given the opportunity to participate in cognitively stimulating activities. Indeed “early” adoption of these activities has been linked to improved outcomes. On this basis, it could be argued that these types of interventions should be extended to people with prodromal AD (IWG) or MCI due to AD (NIA-AA). Research should therefore be carried out to check the benefit of existing interventions in this population and to guide the development of specific interventions for this group.
On the other hand, there may be existing therapeutic interventions or treatment options available through health and social care services for people living with AD dementia that are simply not suitable for people with a clinical classification of prodromal AD (IWG) or MCI due to AD (NIA-AA). The characteristics of people in this classification group is likely to be different from those of people currently receiving a diagnosis of AD dementia, particularly in relation to age and functional ability to engage with different interventions or activities. We can learn lessons from examining the experience of people with early onset dementia and the suitability of some of the health and social care services to which they have been referred. Typically, services for people with dementia are integrated into older people’s services (Haase, 2005; Alzheimer’s Society 2006, 2007). Referral to such services that do not meet the needs of younger people can have a negative impact on a person’s sense of self and identity (Harris & Keady 2009). This suggests that services should be developed for people in the newly classified groups, which are based on a careful analysis of their needs. Existing services which were designed for people with other conditions should not be offered unless they also correspond to those identified needs.
Existing and future possible drug treatments
Introducing a clinical classification of prodromal AD (IWG) or MCI due to AD (NIA-AA) in practice raises issues regarding the suitability, availability and efficacy of pharmaceutical treatments for this group. A survey of over 4,000 respondents, 2,889 with experience of at least one medication used for the treatment of dementia, indicated that three quarters of respondents, including people with dementia, carers and professionals, were largely positive and felt that dementia medication was to some degree beneficial (Alzheimer’s Society 2004). This shows an expectation that pharmaceutical treatment options are part of the care pathway. However, at present, most anti-dementia drugs are not licenced for people without a diagnosis of AD dementia. From a research perspective, prodromal AD (IWG) and MCI due to AD (NIA-AA) are good targets for testing new pharmaceutical interventions, so in the future treatment options may also be developed which are suitable for earlier phases of AD.
Increasing interest in developing drugs for at-risk groups may eventually lead to new drug interventions which are successful for these groups. However, given the potential side effects of the medication, the risks of treatment will need to be carefully balanced against the potential benefits (Molinuevo et al. 2016). This benefit/risk assessment needs to be done differently for people with prodromal AD (IWG) or MCI due to AD (NIA-AA), who already have some symptoms) than for people who are asymptomatic but at risk of AD. Moreover, given that the latter exhibit no clinical symptoms of disease, it would have to be ascertained whether treatment should be afforded within the field of health and social care, or if this type of intervention falls in the domain of public health.
Coping with diagnosis and the need for post-diagnostic support
Existing research indicates that readiness and preparedness to deal with a diagnosis impacts on patient outcomes. Preparing people to adjust to a diagnosis and supporting them through this transition is one of the goals of health and social care services. Negotiating this transition relies on the quality of the treatment provided, as well as on factors related to the individual. In response to a diagnosis of dementia, people tend to fall on a continuum between taking a self-maintaining stance - whereby they try to normalise and minimise their difficulties, or a self adjusting stance in which new information about their cognitive status and difficulties that may lie ahead are integrated into their new sense of self (Clare 2003). These stances are related to different coping styles; self-maintaining to avoidant coping, and adjusting to problem-focused coping. A prodromal AD (IWG) or MCI due to AD (NIA-AA) diagnosis would provide practitioners with opportunities to prepare people better for the transitions associated with the progression along the AD continuum. If good quality informational services are provided at the earliest possible stage, this may facilitate the adoption of a self-adjusting stance with potentially better patient outcomes. Conversely, a prodromal AD (IWG) or MCI due to AD (NIA-AA) diagnosis could potentially encourage individuals to be hyperaware, seek to identify problems that do not exist, and increase anxiety and psychological distress; potentially leading to overuse and unnecessary use of health and social care services.
Socio-economic and cultural issues
There is a pervading question concerning resource use and financial burden that arises from introducing new diagnostic classifications. It would be reasonable to assume that introducing categories that identify AD pathology at earlier stages will increase the number of people with a diagnosis or who are deemed at-risk, and thus the cost to services. The World Alzheimer Report (ADI 2010) indicates that the average annual societal costs are USD32,865 per person with dementia. However, the report suggests that early diagnosis of dementia costs USD5,000 per person and that early diagnosis costs are offset by projected future savings from delayed institutionalisation, with savings of USD10,000 per person across the disease course. Introducing the preclinical categories may eventually help us to identify people who are likely to progress to AD dementia at an earlier point, and potentially result in savings to treatment costs. However, more research would be required to investigate the cost/saving benefit of the classifications.
A further consideration regarding the impact of the new classifications relates to cultural variations in the use of health and social care services. Owing to cultural variations in the perceptions of dementia, socioeconomic issues and social conventions, there already exist variations in the extent to which people from Black, Asian and Minority Ethnic (BAME) communities access health and social care services. In addition, we need to ensure that services themselves are culturally appropriate for people from different cultural backgrounds. For example, in the UK it is acknowledged that treatment and support services for people from BAME groups are inappropriate and lacking (APPG Dementia 2013). The provision of appropriate and high quality treatment and services for all cultural groups related to existing classifications of AD needs to be addressed, and a sensitive understanding of the implications for introducing new diagnostic classifications is needed.
Even if today, in 2016, we still do not know the precise causes of AD and how to treat it, recent research has resulted in in-depth knowledge about the underlying changes which occur in the brain many years before the onset of AD dementia. The common lexicon proposed by Dubois et al. in 2010 represents an important step towards helping ensure that researchers literally speak the same language and permitting the production of a body of comparable research findings. Although the lexicon has been helpful in providing clarity, the new model and related AD-terms continue to evolve, with new scientific publications linked to the new AD model in the pipeline, in parallel with the increase in knowledge and discoveries related to AD pathology. Confusion and uncertainty may therefore persist for some time as researchers strive to grasp and consistently apply the new definitions, and lay people, health and social care professionals, the media and policy makers attempt to interpret their personal and societal significance and implications.
Promoting understanding and respecting autonomy
Although the preclinical AD classification is not a clinical diagnosis and is currently limited to the research domain, the disclosure of risk status may nevertheless have an impact on people’s lives. The difference between a research classification and a clinical diagnosis may be unclear to some participants, especially if such information is provided by a person whom they perceive as a medical doctor. As many biomedical researchers are also healthcare professionals, the boundaries between research and clinical practice may at times be somewhat blurred. Despite strict adherence to relevant guidelines for good clinical and ethical practice in human research, some research participants still confuse medical care with research (e.g. assuming that the aim of a study is to improve their own health rather than to produce generalisable knowledge). The term “therapeutic misconception” was coined by Appelbaum and colleagues in 1982 to describe the failure to appreciate the difference between research and treatment. It remains a challenge for researchers to ensure that participants fully understand the information they provide.
As mentioned earlier in relation to diagnosis, people may interpret information about the new definitions of AD within the framework of their current knowledge about AD and dementia. Consequently, there is also a risk, in the context of research, that they might intrepret categories such as preclinical AD, prodromal AD (IWG) or MCI due to AD (NIA-AA) as forms of dementia. This may have ethical implications linked to informed consent (i.e. it could not be considered as informed if based on a misunderstanding). As not all participants taking part in preclinical studies will develop AD dementia, Molinuevo et al. (2016) suggest using the term “asymptomatic at risk for cognitive impairment” when communicating with research participants. Whilst not an everyday term that people are familiar with, it emphasises an at risk status of a condition that people are perhaps less likely to associate with AD dementia. The impact of providing information about a risk of AD dementia is currently being explored by the European Prevention of Alzheimer’s Dementia (EPAD) study, which amongst other things is also exploring whether reactions to the disclosure of risk status differ depending on the nature of the risk (e.g. linked to genetics, lifestyle factors or biomarkers).
The right not to know must be equally respected. It would be unethical to thrust information about AD status on people, on the pretext that they must be informed, especially in the absence of support and treatment. Consequently, a disclosure process similar to genetic counselling is often used by researchers whereby the possible significance of information they might receive (e.g. about being in a pre-dementia stage of AD) is explained to participants before they are asked whether or not they would like to receive such information.
Possible impact of communication of risk status
For some research participants, the disclosure of risk, and frequent reminders about that risk throughout the duration of the study, may plunge them into a period of uncertainty (Molinuevo et al., 2016). Awareness of risk status may even affect the performance of some research participants on certain tasks through stereotype threat (Molinuevo et al. 2016). Stereotype threat means that a person who has been labelled as having a particular attribute, for which there is a negative stereotype, may feel apprehension about confirming that stereotype when performing a task (Steele 1997). Molinuevo et al. (2016) describe a study in which people who were informed that they had a risk of developing AD dementia had a poorer score on cognitive tests compared to others with the same risk who had not been informed about their risk status. The issue of stereotype threat may also be relevant in everyday life (e.g. in relation to managing personal finances and administrative matters). However, there are individual and environmental factors which affect whether and if so how people are likely to be affected by stereotype threat, and it is important to explore methods to mitigate this effect (Singletary et al. 2009).
People who believe they are at increased risk of AD dementia may be particularly interested in participating in studies into the prevention of AD dementia. When designing measures to protect research participants who are at risk of AD dementia from harm, it is important that researchers consider the possible impact of knowledge about risk status and the timing of disclosure of that risk on the decision whether to participate or remain in a study. Providing information which may cause concerns about health (i.e. disclosure of risk status) whilst offering a possibility to do something which might be perceived as reducing the likelihood of it occurring (e.g. based on the therapeutic misconception mentioned earlier) could be experienced as a subtle form of persuasion/pressure.
Privacy and confidentiality issues
The acquisition of information about the risk of developing AD dementia also raises ethical issues linked to privacy and confidentiality. Schicktanz et al. (2014) emphasise the need to pay attention to questions of disclosing, storing and passing on such information. They suggest that measures need to be taken not only in relation to potential psychological distress but also to familial issues and possible social discrimination. According to Molinuevo et al. (2016), legal safeguards to protect biomarker data provided by research participants are currently inadequate. They suggest that participants should not hesitate to ask researchers what measures are being taken to protect their data.
Sharing potential risks and benefits
In keeping with the principles of equity, solidarity and autonomy all members of society should, in the context of research, be protected from harm but also have an equal opportunity to participate and potentially benefit from a particular study (either personally or through the satisfaction of being able to do something for the good of future patients and society). However, by extending the definition of AD to incorporate an at-risk stage and a pre-dementia stage such as prodromal AD (IWG) or MCI due to AD (NIA-AA), in addition to the dementia phase, participants may eventually be involved in AD research for longer periods of time than in the past (Molinuevo et al. 2016), either by participating in several consecutive studies or by being involved in longitudinal studies (studies which involve participation over a fairly lengthy period of time). If the same people are constantly solicited (as they have been identified and it is often difficult to recruit a sufficient number of participants), they may be disproportionately exposed to research, sometimes with risk. Those who have specific biological characteristics of particular interest to researchers (e.g. a certain gene known as ApoE4) may be particularly sought after for further studies.
This issue is not limited to biomedical research or clinical trials. Researchers conducting other types of research are also increasingly interested in involving participants with a greater likelihood of developing AD dementia in the course of their study in order to provide evidence for or against their research questions or hypotheses. In providing a possibility to identify groups of people with very specific biological and clinical criteria, which are more likely to shed light on their hypotheses, the new AD criteria may contribute towards minimising the unnecessary involvement of some participants (Molinuevo et al. 2016).
The new AD model represents an important step towards developing measures which might eventually reduce dementia by delaying or preventing its onsent and hence improving the quality of life of many people. According to Leibing (2015), this reflects a more general trend towards healthy and active ageing, incorporating the identification of at-risk groups and a focus on prevention for a range of medical conditions. However, this model and the AD-related terms also have implications for equity in research on quality healthcare. They may lead to a disproportionate focus on preventing or delaying the onset of AD dementia and less interest in and hence reduced funding for research aimed at finding more effective therapies (or better care) for people who already have AD dementia. This would mean paying less attention to improving the treatment of one group of people and more attention to improving the treatment of another group of people, based on a new model and hypotheses which have yet to be confirmed. On the other hand, it is important to understand that the new emphasis on the earlier stages in the disease process is partly in response to disappointing results in the development of new drugs for AD dementia. Many researchers now believe that it may be necessary to start therapies much earlier in the disease process, before the onset of dementia (Winblad et al. 2016).
Last Updated: Thursday 16 February 2017